difNLR() function was
created.THIS IS A CRAN VERSION
match argument of the difNLR(), difORD(), and ddfMLR() now handles
a numeric matrix where each column represents the matching criterion for
each item of Data.
match argument of the difNLR(), difORD(), and ddfMLR() now can take
also values of "restscore" and "zrestscore" representing total score
without item being currently tested and its standardized version,
respectively.constraints is now used in the startNLR() function when
computing starting values for the difNLR() function.difNLR(), difORD(), and ddfMLR() functions was
updated.difNLR(), difORD(), and ddfMLR() functions was
updated; specifically for match, anchor, and purify arguments. New
output match.name describing name of the matching criterion for plotting
was added.THIS IS A CRAN VERSION
It includes versions 1.5.1-2 - 1.5.1-4
plot() S3 method for the ddfMLR() was fixed. Categories
are now correctly ordered for lines and points.difNLR() and difORD() functions was updated.genNLR() now handles parameters for dichotomous items as matrices with
a single column.difNLR function
were fixed.
predict.difNLR was
fixed, fixing also issues in the plot.difNLR.df output of the difNLR() function was
fixed. Degrees of freedom are now correctly returned for the LR and Wald
tests.purify = TRUE or anchor argument is used, matching criterion is now
correctly computed in the plot() and predict() S3 methods for the
difNLR() function.fitted() S3 method for the difNLR was fixed.start input in the difNLR() function was updated.genNLR(), difNLR(), difORD(), and
ddfMLR() functions.difNLR(), difORD(), ddfMLR(),
formulaNLR(), and genNLR() functions.anchor was added into the difNLR(), difORD(), and ddfMLR()
functions, specifying anchoring items from the anchor argument. This also
excludes DIF items when item purification is applied with the purify = TRUE
argument.NLR() when method = "irls" fixed. Parameters are now labelled as
b0-b3 and estimates are correctly printed with the coef.difNLR() S3 method.
coef.difNLR() S3 method fixed when only one parameter is
estimated. Thanks to Jan Netik.coef.difNLR() S3 method fixed when multiple items did not converged.difNLR() handling items that failed to converge and their estimates.THIS IS A CRAN VERSION
It includes versions 1.5.0-1 - 1.5.0-2
formulaNLR() when type = "b" fixed.
estimNLR() when method = "plf" fixed.NLR() calculating SEs when no item is converged fixed.testthat package has been started.NLR() function was fixed.
coef.difNLR() when convergence issues are present were fixed.
Thanks to Jan Netik.NLR() and difNLR() functions were updated.
coef.difNLR() was updated.THIS IS A CRAN VERSION
predict.difNLR() was fixed.
startNLR() was fixed."em" and "plf" were added for the method argument in the
estimNLR() function to estimate item parameters with either the EM
algorithm or algorithm based on parametric link function (PLF). "plf" is now
default option. This is also the default option for the NLR() function.
parameterization argument of the formulaNLR() and
startNLR() function were updated (renamed).parM0 are returned with this parameterization.constraints were added into the startNLR() function."likelihood" option for maximum likelihood estimation in the estimNLR()
function was renamed to "mle".
estimNLR() function were extended and improved.THIS IS A CRAN VERSION
plot.ddfMLR() now correctly plots ordinal data.
test = "W" was fixed for the difNLR() and NLR() functions.difNLR() and NLR() functions.startNLR now handles missing values. Returns error when not enough complete
observations are provided.
ggplot2 plotting methods were updated to follow changes in
the ggplot2 package.ggplot2 plotting methods were
updated.ggplot2 v.3.4.0 is now imported.difORD() and ORD() functions were updated. Now using
the Anxiety dataset from the ShinyItemAnalysis package.class handling was updated.It includes versions 1.3.7-1 - 1.3.7-3
parameterization = "logistic" was fixed in formulaNLR()
function.difNLR(), NLR(), and estimNLR() functions.
coef.difNLR(), coef.difORD(), and coef.ddfMLR() methods now
include delta method for IRT and logistic parameterizations.coef.difNLR(), coef.difORD(), and coef.ddfMLR() methods now
include calculation of confidence intervals.estimNLR() function is now unified via print() method.predicted.difORD() to compute predicted values for
difORD object was implemented.plot.difNLR() fixed.THIS IS A CRAN VERSION
Data in ddfMLR() to fix bug
when plotting.method = "nls" was
implemented into the vcov() method for the output of the estimNLR()
function.
difNLR() function.method = "nls" was
implemented into the difNLR() function via an argument sandwich = TRUE.THIS IS A CRAN VERSION
difNLR()
function was fixed.THIS IS A CRAN VERSION
difNLR()
was fixed.
difNLR() for non-converged items including naming
of parameters was fixed (Reported by Jan Netik).NLR(), function
gives warning and NA values for covariance matrix and vector of standard
errors are returned.predict.difNLR() method.
difNLR().plot.difNLR(), plot.difORD() and plot.ddfMLR()
were removed. Change of colours/linetypes/shapes/title can be managed using
standard ggplot2 syntax.plot.difNLR() now offers possibility to turn off drawing of empirical
probabilities using argument draw.empirical = FALSE.plot.difNLR() now offers possibility to plot confidence intervals for
predicted values as offered in predict.difNLR() using argument
draw.CI = TRUE.startNLR() were improved for score as
matching criterion using argument match.plot.difNLR(), plot.difORD() and plot.ddfMLR() were unified.
plot.difORD() and plot.ddfMLR() were changed to blind-color
friendly palettes.THIS IS A CRAN VERSION
plot.difNLR() was fixed.THIS IS A CRAN VERSION
It includes versions 1.3.0-1 - 1.3.0-6 and following changes:
plot.difNLR() now correctly uses matching criterion when item purification
is applied.markdown.
MLR() function now returns correct value of log-likelihood for
alternative model.NLR() function was set
to "all" instead of "both".
Data in difNLR() function can be also a vector now.MLR() was fixed for binary data and IRT parametrization.
print.difORD() method.plot.ddfMLR() was fixed for binary data.ddfORD() was renamed to difORD().genNLR() with an option itemtype = "nominal" returns
nominal items as factors with levels presented by capital letters.
plot.ddfMLR() was updated to show P(Y = option) instead
of option alone.NLR() estimation.item for S3 methods of difNLR class can be now
name of the column in Data.plot.ddfMLR() and plot.ddfORD() were updated.difNLR() function was set
to "all" instead of "both".styler was used to improve formatting of the code.
ShinyItemAnalysis was added into Suggests.estimNLR() was improved.plot.ddfORD() is now correctly displayed.THIS IS A CRAN VERSION
It includes versions 1.2.3 - 1.2.8-4 and following changes:
print.difNLR()
print.ddfORD() and print.ddfMLR().
plot.ddfORD() uses anchor items.plot.ddfORD() now works when Data is factor.genNLR() now generates ordinal data using adjacent category
logit model with argument itemtype = "ordinal".plot.ddfORD() now works when items have different scales.
anchor is now used for calculation of matching
criterion in function ORD().ddfORD().logLik.ddfMLR() now works properly.plot.ddfORD() and plot.ddfMLR().plot.difNLR() can be
changed with group.name argument.difNLR(), ddfMLR(), ddfORD(), MLR(), and ORD()
functions were updated.ddfMLR() function with
argument parametrization. SE calculated with delta method.
plot.ddfMLR() can be
changed with group.name argument.ddfORD() function was renamed. Now ddfORD().
ddfORD() function with
argument parametrization. SE calculated with delta method.plot.ddfORD() can be
changed with group.name argument.ddfORD() was updated.
ddfORD() was added.item in S3 methods for difNLR(), ddfMLR(), and ddfORD()
was fixed.plot() outputs for difNLR(), ddfMLR(), and ddfORD() functions
were unified.plot() for ddfORD() was implemented.AIC(), BIC(), logLik(), coef() for ddfORD() were implemented.
AIC(), BIC(), logLik(), residuals() for difNLR() and ddfMLR()
objects now handle column names as item argument.coef() for difNLR and ddfMLR objects were updated. Their now
includes arguments SE (logical) to print standard errors and simplify
(logical) whether list of estimates should be simplified into a matrix.ddfORD() and ORD() for DDF detection for ordinal data
with adjacent and cumulative logistic regression models were added.
Output is displayed via S3 method print.ddfORD()ddfMLR(), MLR(), and difNLR() were updated.plot.ddfMLR() now handles also binary data.
ddfMLR() returns consistently "No DDF item detected" when no DDF
item was detected.plot.ddfMLR() was improved for displaying
more smooth curves.THIS IS A CRAN VERSION
It includes versions 1.2.1-1 - 1.2.1-3
AIC(), BIC(), logLik() of ddfMLR() are now item specific.difNLR()
NLR()
initboot = FALSE now works properly.difNLR():
ddfMLR():
THIS IS A CRAN VERSION
It includes versions 1.2.0-1 - 1.2.0-7
start in difNLR() function is now item-specific. The input is
correctly checked.
difNLR() and NLR() functions.constraints in difNLR() function is now item-specific.print() method for difNLR class.
difNLR class are now properly described, especially,
plot.difNLR() and predict.difNLR().
difNLR() documentation was improved.difNLR can now properly handle items with convergence
issues.
NLR() now detects DIF correctly with F test.print(), plot(),fitted(), predict(), logLik(), AIC(), BIC()
and residuals() for difNLR class now handles item specific arguments
(model, type and constraints).
residuals for difNLR class now uses argument item.difNLR was fixed and improved.
NLR().NLR().difNLR class can now handle convergence issues.difNLR-package was updated.
plot() and residuals() for difNLR was slightly improved.logLik() for difNLR now returns list of logLik class values.startNLR() now handles item-specific arguments (model and
parameterization). Its output is now in the form of list. It can be
simplified with argument simplify into table when all parameterizations
are the same.
NLR() now handles item-specific arguments (model, type and
constraints).difNLR() now handles item-specific arguments (model, type
and constraints).estimNLR() in NLR() are now properly named.
formulaNLR() was fixed.formulaNLR() and estimNLR() were improved.genNLR() can now also generate nominal data based on
model specified in ddfMLR().
parameters in genNLR() is no longer applicable.a, b, c, d were added into genNLR() as parameters -
discrimination, difficulty, guessing, inattentiongenNLR() can now also generate different underlying
distributions for reference and focal group with arguments mu and sigma.estimNLR() to estimate parameters of NLR models
was added. This function uses non-linear least squares or maximum
likelihood method.
NLR() now uses estimNLR() for estimation of models
parameters.difNLR() can now estimate models parameters with also maximum
likelihood method.estimNLR()
function. This option is not fully functional.plot() for ddfMLR class in matching criterion was fixed.
NLR() was fixed. User-specified starting values are now available.startNLR() was fixed. Function runs even if there are not unique cuts for total scores/match.estimNLR() was fixed.NLR() was done.
NLR() function was fixed.
match argument in difNLR() function was fixed.Data in difNLR() function was fixed.startNLR() function was improved.
ddfMLR() and MLR() can now handle also total score
or other user-specified matching criterion.plot() for class ddfMLR can also handle total score
or other user-specified matching criterion.checkInterval() was added.
difNLR() and ddfMLR().residuals.difNLR() was added.AIC() and BIC() for difNLR class were
updated.plot(), fitted() and predict() for difNLR class
can now handle also other matching criteria than zscore.THIS IS A CRAN VERSION
startNLR() function for missing values was fixed.difNLR() and ddfMLR() functions was
mildly updated and unified.THIS IS A CRAN VERSION
plot.difNLR() was fixed.
constraints arguments in NLR() and formulaNLR()
functions were set to NULL.NLR() function
by startNLR() function.difNLR() function can handle Data with one column.startNLR() now works when match argument is set.formulaNLR() function.NLR() function.startNLR() was mildly updated.ddfMLR() function.
ddfMLR() function.MLR() function.logLik.ddfMLR() function was fixed.
difNLR() was updated.difNLR() function.
difNLR() function.NLR() function.difMedical, difMedicaltest, and difMedicalkey
were renamed. Now they are MSATB, MSATBtest, and MSATBkey.
from Medical School Admission Test in Biology.formulaNLR() was implemented. Function
returns formula for NLR model for 11 predefined models and 4
predefined DIF types to test. Model and DIF type can be also
specified with constraints on parameters a, b, c and d.
NLR() now handles 11 predefined models and 4
predefined DIF types to test. Model and DIF type can be also
specified with constraints on parameters a, b, c and d.startNLR() was edited to return starting
parameters with different parameterization. It was also mildly
changed to correspond to new version of NLR() function.difNLR() can now handle also total score or other
user-specified matching score.constrNLR() is no longer part of the difNLR
package.difNLR() and ddfMLR() functions.difNLR() function.msm package is now used for delta method in difNLR() function.THIS IS A CRAN VERSION
plot.ddfMLR() for non-uniform DDF was fixed.
THIS IS A CRAN VERSION
difNLR() function was fixed.
GMAT and GMATtest were extended by criterion variable which is intended to be predicted by test.coef, logLik, AIC and BIC S3 methods were added for class ddfMLR.plot.ddfMLR() and plot.difNLR() were slightly improved.
difNLR() and ddfMLR() functions.THIS IS A CRAN VERSION
ddfMLR() to detect Differential Distractor Functioning (DDF) with Multinomial Log-linear Regression (MLR) model. S3 methods for class ddfMLR also added - print and plot.
MLR() to calculate likelihood ratio statistic
for detecting DDF with MLR model.difNLR() function can handle 6 generalized logistic
regression models with option model.startNLR(), genNLR() and S3 methods for class
difNLR were changed according difNLR() function. S3 method
coef was created.NLR() and constrNLR() can now calculates DIF
detection statistics and specify constraints for generalized
logistic regression model.difNLR() was edited to response to difR package
and its DIF detection functions.genNLR() was changed to generate dataset from
generalized logistic regression model with 8 parameters.AIC(), BIC(), and logLik() S3 methods added to difNLR().THIS IS A CRAN VERSION
plot for class difNLR was updated.
test in difNLR() function was added. Possible
choices are now F for F-test and LR for likelihood ratio test.alpha was added into difNLR()
function with default option 0.05.GMAT data, its unscored version
GMATtest and its key GMATkey. Scored difMedical data set, its
unscored version difMedicaltest and key difMedicalkey.genNLR() was added to generate scored (binary) data with
model by difNLR.